Fra grænselandet mellem to kulturer: Tragtbægerkultur og enkeltgravskultur i det centrale Østjylland

From the borderland between two culturesThe Funnel Beaker culture and the Single Grave culture in central eastern Jutland Eighty years ago, J. Brøndsted and especially P.V. Glob painted colourful pictures of invading “axe-swinging” nomads from the south who conquered the central and western parts of Jutland. The Jutland Sing­le Grave culture (SGC) was indeed different from the established Funnel Beaker culture (FBC) in southern Scandinavia and part of its background obviously lay in central Europe. The chronological relationship between the FBC and the SGC has been intensely debated and viewed very differently over the years. 14C dates now provide us with a reliable temporal correlation between the two cultures. The latest part of the FBC – MNA V or the Store Valby phase – roughly dates to between 3000 and 2600 BC, while the SGC began around 2800 BC and lasted until 2350 BC, at the transition to the Late Neolithic. The end of the FBC corresponds to the transition from E. Hübner’s SGC period 1 to period 2, or the earlier part of the Bottom Grave period in P.V. Glob’s system.Through several years, I have collected a comprehensive set of data about the Neolithic from a 640 km area in central eastern Jutland, around and north of Horsens Fjord (fig. 1). During period 1 of the SGC, the culture existed exclusively in central and western Jutland, as evident from a distribution map of its graves based on E. Hübner’s investigations (fig. 2). Within the study area, the SGC was only present in the northwest corner, while the contemporary parts of the FBC were bound to the coast. The aim of this article is partly to elucidate the relationship between the SGC and the FBC within the study area during the period when they existed side by side, and partly to look at what happened when the FBC disappeared. To understand why the FBC became bound to the coast, and why it finally collapsed, I will also examine the internal development of this culture prior to the appearance of the SGC. The large, shallow, funnel-shaped Horsens Fjord with its characteristic islands is central to the study area. In its innermost part, a narrow passage – Stensballe Sund – connects the main fjord with a minor, brackish inner fjord. During the Stone Age this was part of a much larger inner fjord with a depth around 13 m and extending 4-5 km inland, but sediments from the rivers have now filled it in. Following the general pattern of sea-level rise and land upheaval after the Ice Age, we would expect to find the Late Mesolithic Ertebølle settlements 1-2 m above current sea level, but they actually lie 1 m below. The reason is that Horsens Fjord is a geological subsidence area, which has been active for millions of years. This has caused the sea level to remain stable since the Mesolithic, though it has fluctuated in connection with the various transgressions and regressions.The FBC in eastern Jutland prior to 3000 BCThere are relatively few finds from the study area dating from the earliest part of the Neolithic (EN I – 3900 to 3600 BC), but we do have shell middens along the coast, most notably at Norsminde, as well as a number of small short-term settlements located away from the coast. One of the latter is the well-preserved and fully excavated site at Mosegården. A causewayed enclosure at Aalstrup, four graves and four depositions in marine environments also date from the period.Towards the end of the Early Neolithic and through the earlier part of the Middle Neolithic FBC (EN II-MNA II (II/IV) – 3600 to 3000 BC), the quantity of finds increases radically (fig. 4). Especially at the head and along the north side of Horsens Fjord the number of settlements is significant, many of them well documented through excavations. In addition, there are three causewayed enclosures, numerous graves – mostly megalithic tombs – and many depositions in freshwater and, especially, marine environments.The numerous finds give us a detailed picture of the settlement location and structure. The settlements were clearly bound to the coastal zone, especially by Horsens Fjord. On the north side of the fjord, the finds tend to cluster around the three known causewayed enclosures – from east to west: Aalstrup, Toftum and Bjerggård – and given the major cluster at the head of the fjord, the site of Aarupgård was probably a causewayed enclosure too. The number of inland finds is limited, and those occurring in the northwest part of the study area all appear to be of an early date. The magnitude and dates of the settlements along the coast, the mixture of both early and late tombs in the clusters around them and the repeated depositions evident in front of the tombs emphasise the permanency of the occupation.In 1941, J. Iversen suggested that pollen diagrams reflected the introduction of agriculture – “landnam” or land taking – in southern Scandinavia. He identified three phases: forest clearance, slash-and-burn agriculture in the clearings and forest regeneration. 14C dates now show that the three phases represent three different forms of forest-based agriculture, with the first phase dating to between 3900-3600 BC, the second phase to between 3600-3000 BC and the third phase to between 3000-2600 BC. The first, almost invisible, phase in the pollen diagrams fits well with the small, scattered short-time settlements we find during EN I. The same is true of the second phase, when the clusters of settlements through EN II – MNA II formed the basis for a permanent slash-and-burn rotation with clearance, burning, cultivation, grazing and forest regeneration, primarily by birch. Two other sources that highlight the land-use patterns are plant macrofossils and animal bones recovered from the settlements. Figure 5 shows the relative distribution of the various types of cereals through the Neolithic, based on the macrofossil evidence, while figure 6 shows the relative distribution of cattle, sheep/goat, pigs and hunted mammals throughout the FBC, based on the archaeozoological evidence. We find that wheat, particularly emmer, totally dominated during the EN and the beginning of MNA, although during the latter the proportion of barley, almost exclusively naked barley, grew to a third. At the beginning of the Neolithic the “forest feeders” – pigs and hunted mammals – dominated over the “meadow animals” – cattle and sheep/goat. During MNA, the proportion of cattle grew considerably, while that of hunted mammals decreased.In addition to the settlement and land-use patterns, our knowledge of the structure of society relies on monumental tombs and causewayed enclosures, both of which appeared around 3800-3700 BC. The earliest burial structures were long barrows with wooden chambers, many of which lay on former settlements. Judged from the unfortunately few bones preserved from wooden chambers, these tombs appear not to have been reserved for high-ranking individuals. We find examples of multiple burials in a couple of chambers and burials of children are also evident in a larger number of cases. In some of the wooden chambers, however, the nature of the grave goods does imply a form of social stratification. This is clearly evident in the case of the Rokær tomb, located within the study area, which is 14C dated to between 3400-3300, at the transition from EN to MNA. The chamber contained two thin-butted flint axes, a very special polished blade knife and numerous amber beads. The latter are of the same types as those deposited in a pot, together with copper ornaments, at the presumed causewayed enclosure at Aarupgård, only 1 km from Rokær. During EN II, stone replaced wood and dolmens were built in ever-increasing numbers and monumentality. In MNA I, the chambers grew in size, and narrow passages, which could be blocked with door stones, were added. But then, at the end of this period, the building of new tombs ceased. Whereas a few pots were deposited at the eastern end of the long barrows during the EN, at the beginning of MN A the deposition of large numbers of pots in front of the entrances to the tombs became the rule; a custom that continued after the building of new tombs had ceased. The excavations of three megalithic tombs in the study area, Stenhøj, Nørremarksgård and Grønhøj, have revealed new interesting details of the customs of deposition.With the increased building of monumental tombs, we find that these tended to cluster around the settlements (fig. 4). This is especially true at the head of Horsens Fjord, where we also find that settlements and tombs were placed in separate areas. The graves were in contemporary use, in the sense that activities in the chambers and at the entrances occurred at intervals of 10-20 years, as suggested by the analysis of the depositions at Nørremarksgård.The general view of a causewayed enclosure in southern Scandinavia is based on the Sarup I enclosure with its complex palisade system and double row of ditch segments aligned with the palisade. Very few causewayed enclosures were, however, like this. Most had one or two rows of often irregular ditch segments. Toftum, in the study area, is a good example of this. It originally consisted of a single row of ditch segments of varied shape and formation history. Some were short and deep, while others were long and shallow; some remained open, partially silting up before being backfilled, while others were backfilled shortly after they had been dug. There is no datable material from these early stages. Later, apparently after a large settlement was established in the eastern part of the enclosure in EN II, a second row of inner ditches was added along the western periphery of the enclosure. These ditches were backfilled and recut several times, with many of the recuts containing ritual depositions and increasingly large amounts of settlement debris too. Recuts occurred in a few of the ditches in the outer row at this time as well.It is a widely held opinion that causewayed enclosures were permanent central sites for social interaction and trade. I have difficulty seeing this, given the pattern of activity we find associated with the ditch segments. I believe instead that the causewayed enclosures were probably ritually conditioned, most likely in connection with death rituals, as indicated by preserved human bones in ditch segments at some of these monuments. Furthermore, the numerous depositions of pottery in the ditch segments constitute a parallel to the depositions undertaken in front of the megalithic tombs. These emerged at the same time as those in the ditch segments began to disappear. The causewayed enclosures were, as it has been expressed, “villages of the dead souls”, and the row of ditches formed the borderline between the dead and the living. From late in the EN into the early MNA, the villages of the dead gradually became the villages of the living. The souls moved to the chambers of the megalithic tombs, while actual settlements began to appear inside the causewayed enclosures, as seen at Toftum.The FBC and the SGC in eastern Jutland between 3000 and 2600 BCThe FBCThe FBC changed profoundly in its final phase between 3000 and 2600 BC. The settlements were fewer and moved closer to the coast, and we know of only two burials. Depositions in marine environments, on the other hand, proliferated (fig. 7). What the settlements lost in numbers, they gained in size and temporal extent as demonstrated by investigations at several sites. With large, permanent populations, exploitation of the resource area around the settlements became intensified. The importance of cattle (fig. 6) and barley (fig. 5) increased. A study of flint sickles has shown an increased use of these towards the end of the FBC, as well as the introduction of a new harvesting technique adapted to the handling of naked barley.In the third phase of the “landnam”, contemporary with this part of the FBC, we find a marked increase in the hazel curve, which reflects an increase in the use of wood pastures for cattle grazing. This practice is well known historically, where the idea was to find a balance between freestanding trees, especially hazel, and grass cover between the trees. The trees are kept at bay by shredding and coppicing, and the cut branches are stored for winter fodder. The high incidence of hazel was also advantageous for pigs. It is likely that cattle were kept and fed on permanently cleared fields close to the settlement during the winter. These fields could then have been used for cereal growing during the summer, taking advantage of the cattle dung. We know that manuring took place in the FBC, but currently it is not possible to qualify this.The SGCDespite their scarcity, the few finds from period 1 of the SGC in the study area show a clear distribution pattern (fig. 8). Within a limited area to the northwest, we find three probable settlements, 13 graves and five stray finds of battle-axes, probably all from graves. From the coastal region of the Late FBC, we have two battle-axes and an amber bead deposited in fresh water and five stray finds of battle-axes.Our knowledge of early SGC land use is almost non-existent. Faint traces of settlements allow us to conclude that the settlement units were small and short-lived. Pollen diagrams from lakes and sediments below barrows in central Jutland constitute our best source. These show a marked reduction in forest from around 3000 BC, combined with burning and the formation of heathland. The diagrams from below barrows reflect heavy grazing in open areas dominated by heather. Imprints of cereal grains in pottery show that both wheat and barley were cultivated, and further east in Jutland we have pollen of barley associated with two barrows and ardmarks from ploughing below a further two barrows, one of which is in the study area.The most eye-catching aspect of the SGC, seen from a FBC point of view, was the re-establishment of the individual in the social universe. In burials, the individual was at the centre, and accompanied by grave goods that signalled a society where gender differentiation was a key issue. At first glance, you might think that it was a society with gender equality and without social ranking between individuals and families, but it was probably not so simple. The ratio between women and men in the early burials was 1:17, and the burials themselves were far from simple. We often see supplementary constructions, in the form of ring ditches and palisades, surrounding the coffins, and in some cases mortuary houses. Major efforts were associated with the burials, and the mortuary houses indicate prolonged rituals. You would expect the most complex burials also to contain the richest grave goods. In some instances, this was the case, as at the Gantrup grave in the study area. In others, however, the only gift in the grave was a flint blade, like the one from Sjørup between Holstebro and Viborg (fig. 15c).The borderlandThrough a 200-year period, from 2800 to 2600 BC, the FBC maintained dense and stable settlement along the coast in the study area, especially around Horsens Fjord (fig. 7). At the same time, we find the SGC in its northwest corner (fig. 8), where it constituted the culture’s easternmost outpost in this part of Jutland during period 1 (fig. 2). We have no finds in between, suggesting the existence of a no-man’s-land. Nevertheless, the finds recorded within the two settled areas give clear indications of mutual contact. From a typical early SGC grave at Højvang (no. 1 in fig. 8), in addition to a battle-axe dating to period 1b (fig. 9a), we have two flint axes and a flint chisel (fig. 9b-d), all of which are typical FBC products, very different from those produced by the SGC. This presence of FBC flint axes in a SGC grave is not unique. Another good example further north in eastern Jutland is a grave at Refshøjgård, and a look at the finds listed in E. Hübner’s catalogue shows many other examples.In the coastal zone, we have three freshwater depositions, comprising two battle-axe fragments and an amber bead. Especially the latter is noteworthy (fig. 10a; no. 2 on the map fig. 8): It is a disk-shaped bead with a central perforation, exclusively found at the waist region in male burials of the SGC. Deposition in water was a standard ritual in the FBC, but unknown in the early SGC. In addition to these finds, there are five SGC battle-axes from the coastal zone, all stray finds. Two of these are of particular interest (fig. 10b-c; nos. 3 and 4 on fig. 8). They are both of type A3; one is in prime condition, while the other is battered and has been intentionally reshaped with one shoulder pecked away. Together with another stray find of a type A3 battle-axe to the northwest, these are the earliest datable SGC artefacts we have from the study area, yet two of them were found in FBC territory. The last find from the coastal zone is of an entirely different kind (no. 5 on the map fig. 8). This is the copper disk from Rude (fig. 11 b). It was found in a stone cist in 1894, tied with copper wire to the wrist of the deceased (fig. 11a), and was published by K. Randsborg in 1970 as an Early Neolithic disk. In 1976, I realised that a couple of scheduled stone cists were identical to those mentioned in the museum registers, and over the next two years, I had them and their surroundings excavated. As the stone cists proved to lie in an Early Neolithic long barrow, the problem was solved and the case closed – until a 14C sample from skeletal remains from the cist containing the copper disk indicated a date of between 3000 and 2600 BC. The copper disk was obviously not Early Neolithic, and new origins were sought. Suggestions have ranged from the Globular Amphora culture to the Corded Ware cultures, first in Bohemia and later in western Switzerland.A culture disintegratesOver a 400-year period between 3000 and 2600 BC, the FBC has the appearance of a stagnant culture in the study area which, apart from the depositions in marine environments, was notable only for large settlements with very coarse pottery and high-quality flint axes. If we look at other parts of southern Scandinavia, we find a somewhat different picture, with cultural changes partly inspired by other cultures. On the islands to the east, the use of the megalithic tombs did not decline to the same extent as in eastern Jutland, and decorated pottery, frequently found in the chambers, remained in use for longer. This is especially evident on the southern islands of Lolland and Falster, where the pottery, and its use in tombs, was strongly influenced by the Globular Amphora culture (GAC) south of the Baltic Sea. On the island of Bornholm, further east in the Baltic, the pottery also continued to be decorated, making it possible to divide the period into two phases. At the same time, outwardly directed ritual activities continued in association with circular wooden buildings placed within palisade-framed areas that appeared in continuation of the causewayed enclosures. Developments in Scania corresponded to those on Bornholm, with the addition of cultural elements from the Pitted Ware culture (PWC), seen by some as having been absorbed by the FBC and by others as evidence of an independent presence of this latter culture.The impact of the PWC all around the Kattegat is obvious, but a map of the distribution of its characteristic tanged arrowheads and bipolar blade cores is not the same as a map of the area where it had settled. There are, however, many settlements with clear PWC associations along the Kattegat coast of northern Jutland and up into the Limfjord. This is especially true on Djursland, where the final Store Valby phase of the FBC was not present. We find instead settlements displaying a mixture of PWC and FBC elements similar to the situation in southernmost Sweden. The FBC stood strong in northwestern Jutland, south of the Limfjord, and underwent the same developments as in the study area, with depositions of pottery in front of megalithic tombs gradually fading away, ultimately being covered by layers of stones. Contrary to other areas, a new form of grave – the stone-heap grave – appeared during the late part of the FBC. Originally seen as rows of graves in front of mortuary houses, these have now been convincingly interpreted as burials in “wagons” (the mortuary houses) with draught animals buried in the rows of graves. These graves probably reflect direct contact with the GAC, which practised ritual burial of cattle, a connection reinforced by finds of FBC battle-axes showing strong GAC influences in the area. In most parts of southern Scandinavia, the FBC was strongly influenced by the PWC and/or the GAC in its final phase, and it is obvious that its days as a culture were numbered. These influences are not so obvious in the study area, but this was not due to ignorance. There are clear indications of contact with both the PWC and the GAC. Figure 12 shows the distribution of PWC tanged arrowheads and blade cores in the study area. But rather than the presence of the culture in the area, it reflects intensive trade, probably in high-quality flint from Djursland. The finds of a battle-axe deposited in Horsens Fjord (fig. 13a) and an amber disk found near Horsens (fig. 13b) imply direct contacts with the GAC, from where they clearly originated as imports.A new culture emergesThe idea of a common European “A-horizon” of the SGC emerged on the basis of P.V. Glob’s study in 1944. Since then, studies of the various local Corded Ware groups in Europe have shown that such a horizon did not exist. More surprising perhaps, it did not exist in Jutland either. But there can be no doubt that major cultural influences, including migration from the south, shaped the ideological structure of the SGC. The c